2015 Informs Annual Meeting

SD43

INFORMS Philadelphia – 2015

SD41 41-Room 102A, CC Healthcare Capacity and Patient Flow Analytics Sponsor: Manufacturing & Service Oper Mgmt/Healthcare Operations Sponsored Session Chair: Retsef Levi, J. Spencer Standish (1945) Professor of Operations Management, Sloan School of Management, MIT, 100 Main Street, BDG E62-562, Cambridge, MA, 02142, United States of America, retsef@mit.edu 1 - The Impact of Delays in Transfer out of the Intensive Care Unit David Scheinker, Postdoctoral Research Fellow, MIT Sloan School of Management, 50 Memorial Dr, E52-289, Cambridge, MA, 02142, United States of America, dscheink@mit.edu, Sara Dolcetti, Benjamin Christensen, Ulrich Schmidt, Retsef Levi, Peter Dunn Few studies examine delays in transfer out of the ICU. We studied four years of patient flow through six ICUs at a large academic medical center. Over 36% of ICU patients transferring to a general care unit experienced a non-clinical delay of over 12 hours. Each day a patient was delayed added approximately a full day to their total hospital length of stay. These results have direct implications for hospital capacity design, bed assignments, and care processes across units within the hospital. 2 - Using Data Analytics and Systems Modeling to Inform Hospital Obstetrics Capacity Planning Nan Liu, Columbia University, 722 W. 168th. St., New York, NY, United States of America, nl2320@columbia.edu, Linda Green Using a recent large data set that contains all hospital obstetrics units (n=40) in NYC, we demonstrate and validate the use of data analytics and systems modeling for planning hospital bed capacity. We estimate capacity needs based on the probability of delay experienced by patients in getting a bed. Our analysis reveals significant variation in obstetrics capacity utilization in NYC; and shows that the whole city can save $26.5M a year with an appropriate reallocation of obstetrics capacity. 3 - Optimization-driven Framework to Understand Healthcare Networks Cost and Resource Allocation Fernanda Bravo, Assistant Professor, UCLA-Anderson, 110 Westwood Plaza, Los Angeles, CA, United States of America, fbravo@mit.edu, Retsef Levi, Marcus Braun, Vivek Farias Consolidation in the HC industry has resulted in the creation of large delivery networks. Traditional practices in cost accounting, e.g., overhead and labor cost allocation to activities, are not suitable for addressing network challenges. We develop an optimization-driven framework inspired by network revenue management to better understand network costs and support strategic network design and capacity allocation decisions. We report the application of this approach on a real case study. SD42 42-Room 102B, CC Patient Scheduling under Resource Constraints Sponsor: Manufacturing & Service Oper Mgmt/Healthcare Operations Sponsored Session Chair: Hossein Abouee Mehrizi, University of Waterloo, 200 University Avenue West, Department of Management Sciences, Waterloo, ON, N2L 3G1, Canada, haboueemehrizi@uwaterloo.ca 1 - Optimal Mix of Elective Surgical Procedures under Stochastic Patient Length of Stay Hessam Bavafa, Assistant Professor, Wisconsin School of Business, Madison, WI, United States of America, hbavafa@bus.wisc.edu We consider the problem of allocating daily hospital service capacity among several types of elective surgical procedures. Our focus is on the interaction between two major constraining hospital resources: operating room and bed capacity. In our model, each type of surgical procedure has an associated revenue, deterministic procedure duration and stochastic hospital length of stay.

2 - Appointment Scheduling Problem when the Server Responds to Congestion Zheng Zhang, University of Michigan, 1205 Beal Ave, Ann Arbor, MI, 48105, United States of America, zzhang0409@gmail.com, Brian Denton, Xiaolan Xie We describe a stochastic programming model for appointment scheduling that incorporates server response to congestion, i.e., the server increases the service rate as the workload grows. It materially differs from previous studies in the sense that the uncertainty in appointment systems is endogenous with respect to the decision variables. We describe properties of the model, methods to solve it efficiently, and results that illustrate the impact of congestion in practice. 3 - Multi-priority Online Scheduling with Cancellations Van-Anh Truong, Columbia University, 500 West 120th St, New York, NY, 10027, United States of America, vt2196@columbia.edu, Xinshang Wang We study a fundamental model of resource allocation in which a finite amount of service capacity must be allocated to a stream of jobs of different priorities arriving randomly over time. Jobs incur costs and may also cancel while waiting for service. To increase the rate of service, overtime capacity can be used at a cost. This model has application in healthcare scheduling, server applications, make-to- order manufacturing systems, general service systems, and green computing. 4 - Multi-speciality Surgery Scheduling under Hospital Resource Constraints Shrutivandana Sharma, Singapore University of Technology and We consider a surgery scheduling problem, where the decision is to schedule surgeries of two different types over a finite planning horizon. The number of surgeries of each type that can be performed in any period is bounded by the availability of operating resources and the availability of beds. We formulate the problem as a multi-period inventory problem, and characterize the optimal solution. Design, 8 Somapah Road, Singapore, 487372, Singapore, shrutivandana@sutd.edu.sg, Hossein Abouee Mehrizi Chair: Pelin Pekgun, Assistant Professor, University of South Carolina, 1014 Greene Street, Columbia, SC, 29208, United States of America, pelin.pekgun@moore.sc.edu 1 - Resource Pricing in Hospitality Industry Xiaodong Yao, SAS Institute Inc, SAS Campus Drive, Cary, NC, 27519, United States of America, xiaodong.yao@sas.com, Tugrul Sanli, Matt Maxwell, Jason Chen Best Available Rate (BAR) pricing is probably the most important pricing decision for hotels. There are two forms: BAR by Day, and BAR by LOS(Length of stay). In BAR by Day, prices are set for each resource, i.e., a pair of (room type,stay night), and a LOS price is just the sum of prices on each resource. While in BAR by LOS, prices are set for each product, a triple of (room type, arrival date, LOS). In this talk, we discuss several methods for solving the resource pricing problem. 2 - Estimating Revenue Variance in the Pricing Models Darius Walczak, Principal Research Scientist, PROS Inc., 3100 Main Street, Suite 900, Houston, TX, 77002, United States of America, dwalczak@pros.com, David Mccaffrey Variance and other distributional moments are important in modeling risk in optimization problem. They are more challenging computationally than linear load metrics such as load factor. We adopt an approach found in the Markov Decision Process literature to calculate variance of revenue under dynamic policies in single-resource pricing problems. We discuss possible extensions to network problems. 3 - Advanced Behavioral Models in Integer Optimization Shadi Sharif Azadeh, EPFL, EPFL ENAC TRANSP-OR GC B3 444, (Batiment GC) Station 18, Lausanne, Switzerland, shadi.sharifazadeh@epfl.ch, Bilge Atasoy, Moshe Ben-akiva, Michel Bierlaire We are interested in discrete optimization models where supply and demand closely interact (airlines). We propose a general methodology leading to an integrated supply and demand model, based on discrete choice that is linear in its decision variables. We illustrate it with an example where a supplier (such as an airline, or a chain of movie theaters) needs to decide to offer some services, and to decide about the price of each slot of the available capacity in order to maximize its revenues. SD43 43-Room 103A, CC Data-Driven Revenue Management Sponsor: Revenue Management and Pricing Sponsored Session

133